Every year, roughly 700 of the 750 million visits that overweight and obese patients make with primary care providers (PCPs) occur without any weight counseling. The main reasons for this are that PCPs are poorly trained to help their patients lose weight and that there are no consistently effective interventions for primary care settings. Though in-person and telephone- based weight control programs have been difficult to disseminate in primary care, online weight control programs are increasingly effective and may lend themselves to be used in these settings. Given the growing number of effective online programs, for obesity and for other conditions seen in primary care (e.g., depression, insomnia) it is important to understand whether these programs can be effective when integrated into primary care and whether they are enhanced by provider involvement. Research on the 5 A's model of primary care behavior change suggests that the most effective, yet least used feature of primary care interventions is arranging follow-up, where providers hold patients accountable to adhering to treatments and achieving specific outcomes. We have created a simple method for integrating an Internet weight control program into primary care settings, by allowing PCPs to monitor their patients' adherence and outcomes and email them pre-written, tailored follow-up messages. PCPs in our pilot work believed that this would help to overcome key barriers to helping their patients lose weight. In this study, we propose to test the impact of integrating an effective automated Internet weight control program into primary care, by recruiting 27 PCPs and 540 of their patients and randomizing them to one of three conditions: A) Brief physician counseling + Usual care, B) Brief physician counseling + Referral and access to a the Internet weight control program and C) Brief physician counseling + Referral and access to the Internet weight control program + brief follow-up email notes of support and accountability from PCPs. We hypothesize that mean weight losses after 12 months will be greater in Condition C v B [5.0kg (SD = 6.0) v. 3.0kg (SD = 4.0)] and in Condition B v. A [3.0kg (SD = 4.0) v. 1.0 kg (SD = 3.0)] and that weight loss will be associated with changes in accountability, logins, eating and physical activity behaviors. Identifying ways, such as proposed, to leverage the doctor-patient relationship to improve patient adherence and outcomes from online programs could have a significant public health impact.